you can read the zero gravity toilet instructions.... and that's exactly why I would watch 2001, and that's exactly why I would buy a DRM-ed blue ray, plus player: To read the zero gravity toilet instructions. In case I happen to sit on one tomorrow.
Of course you are right if you take my words literally, and to that extent. However, if you care to infer the message, it is about the focus with which movies are shot. There is only that much of funding, and in these days, I'm afraid, the average investor is more concerned about resolution and brilliance (of the shots), than in the artistic quality of the undertaking. And all this 'over-technisised' appreciation of the audience will actually lead to movies being shunned because of a perceived lower technical quality, despite of potentially higher artistic quality. I personally have overheard people who refuse to buy any non-BlueRay movie, because "Blue Ray is the future". Content seems to disappear behind technicalities, including for the consumer. And if you please read the message of the OP, I would never want to watch 720x486 NTSC SD interlaced footage EVER again., you might understand my urge to point out what a nonsense this implies. And that one was modded +5, Informative. I was only trying to say, that my primary argument for selecting a movie is its artistic content; not its resolution.
You sound as if you knew what you are talking about, so I take it to try to answer your message. Disparity is the most important factor in 3D perception in the human visual system No doubt. This is why 3D 'works'.
Motion parallax however, cannot be achieved, since hidden content cannot be interpolated. It actually is unknown, eventually to both viewpoints. Even if it is known to one, depth remains unknown.
Head-mounted devices are worse, because for nobody the world doesn't turn when (s)he turns the head. You follow with the shown perspective, I guess. But from where do you get it? Think about a movie: Where do you get the information from, when the viewer turns the head?
Vergence and accommodation (focus) are secondary and always overridden by the other factors; this is a neurophysiological fact Yes, see above. Override, though, does not mean trashed. It remains a sensory effect, that contradicts at least a distance virtually 'close' according to its disparity. -> Headaches.
How does a microlens array induce physical distance (adaptation)? You'd need a set of screens at various distances form the viewer's eyes, and using a shutter mechanism to project specific pixels from a 'credible' distance.
Physiologically, if you inhibit head/eye movement totally, the vision disappears altogether, as you probably know. So we all perform small quantities of those all the time, unconsciously. That's fine for a 2D-display (as I wrote elsewhere in this topic), because that's what we are aware of: a 2D-projection on a plane of finite, if not very limited size. Our brain 'expects' what it gets from watching a picture, or your 2D flat screen. Even a 3D-effect (compiz, e.g.) is nothing but a calculation of virtual distances and structures, projected - visibly - on a 2-dimensional screen. So our eyes get what they expect, with respect to convergence, parallax, focus, etc. Not so, however, if you add real depth/disparity; but none of the others. Tiny, maybe subconscious, movements of eyeballs and/or your head do actually 'explore' the depth; not so in any 'disparity-is-everything'-projection system.
As long as we don't have a projection that makes appear an object in a real 3-dimensional space (what in theory a laser could do), so that the room is real, with a virtual object of real 3 dimensions projected into it, headaches will be the order of the day.
Yes, and no. Your argumentation discards a relevant fact; one that you are probably not aware of.
Black and White photos are a proper representation, or mapping, of a 3-dimensional space on a 2-dimensional plane. Adding colour adds information. The human eyes can be tricked into perceiving a rate of above 16 images per second as 'motion', and an ever higher rate as 'smooth motion'. You add colour to it, everything fine. Over the years, this has been refined, and we can all enjoy coloured moving images without trouble.
Stereoscopy as it is being done, cannot produce a proper mapping. (I gave some initial arguments elsewhere in this thread, so I don't want to repeat myself.) This is why 3D hasn't taken off despite of very early efforts, in red/green, of some generations earlier. The problem is not one of technology, resolution, not even left/right separation. The problem is, and there is plenty of research available if you are interested, that - contrary to the mapping of 3D to 2D - two cameras - even if mounted with the proper interocular distance - cannot map the 3D-impression properly into 2 electronic channels. Therefore, it is physically/biologically impossible to regenerate the original 3D impression with lateral cameras.
1920x1080p Blu-Ray Discs are incredible. I would never want to watch 720x486 NTSC SD interlaced footage EVER again. I work in post production/special fx, so i'm a videophile.
That's one of the troubles with the world of today. Some people get their kicks just from the resolution of the image. Go to any TV-electronics parlor. People will be excited about the crisp picture, the brilliant colours. Whenever I go there, I am infinitely bored with the crappy movies. And then I go home, and watch 720x486 NTSC SD interlaced with an enormous pleasure; Bunuel, Hitchcock, Marx Brothers. Even Kubrick's 2001 is great fun, in PAL. Murnau's Nosferatu (I guess, not more than 300x200 effective resolution) sends more shivers down my spine than Kinski's remake, even if it were in 1080p. Because it is the art; not the resolution that counts.
I'd throw all my mod points at you, if I had or could. Already the next commenter knows close to nothing about stereoscopy, but argues as if he did. You are right from a number of viewpoints (pun!, héhé):
1. At real 3D, when you move your head laterally, you can 'circle' around an object. 2. At real 3D, when you move your head laterally, objects hidden behind other objects become visible. 3. [You didn't mention this one:] Depth perception of the human eyes is done by a combination of biological effects:
- convergence of the eyeballs (like when you watch your finger and bring it closer to your own face, the eyes turn 'inward'
- adaptation of the lenses for a specific distance
The so-called '3D' that can be achieved by two cameras only fulfills one of these features: convergence of the eyeballs (by introducing a lateral offset of the two images on the projection screen). This is why this so-called '3D' gives you some '3D-feeling', but mostly headaches; as the "3D-detection algorithm" in your brain cannot accommodate the incoming information properly; it defeats and contradicts what it has learned throughout your lifespan.
... though I am a GPL fan and against geolocation.;)
If the wine is marked according to the grape, like Merlot or Syrah, or so, fine. If it says something that has locally specific production procedures, or stems from a specific location, I am very much for disallowing fake labels. I think the so-called 'Budweiser' has been mentioned. Except of the name, it has nothing at all to do with the original. Therefore it is misleading, over, and the consumer ought not be mislead. Some wines have specifics, as well as cheese. Like Roquefort, which ripens under a specific situation, in a specific cave. Jerez - actually what made it into Sherry, because the British find it difficult to pronounce Spanish names - has some specifics in its taste that cannot easily been reproduced by inputting some nondescript white wine. Gewürztraminer is another one - though I don't like white wines - in my list. Drank some in Perth, Australia once, and found it unbearable. Just plugging whatever fantasy names there are on the bottle, and off you go? No, as someone who does like red wine, it is even worse for me. A Bordeaux is distinctly different from a Beaujolais Nouveau, and when I buy one, I have a desire to enjoy a distinct taste to which I was looking forward when I bought it. And I want to have this privilege in future as well, even more so. I don't mind at all those crazy and fantastic names for wines that are sold left and right where we live, the 'Night Train', 'Red River' and likewise stuff. But when I buy a St. Emilion, I want a St. Emilion. Anything else is cheating; and I am somewhat astonished that a number of comments in here do not feel likewise. It is not much different from computers: if I buy an intel, I want neither an AMD nor a Via; and if I buy a Via, I don't want an Intel i3. And don't come with all those silly 'double-blind tests'. Should I do likewise, and state: "If you can't distinguish the speeds of two computers, subjectively, (like with the wines), you are not entitled to obtain the type of CPU that you ordered"? When I buy an Intel Core i5 661, that's what I'm entitled to get. Irrespective of my ability to count the pins and measure the clock frequency. And when I buy a Champagne, I want explicitly a product that has been produced according to a very specific and distinct method; not some sparkling white wine with added carbon dioxide.
I suppose if they are brought up in a society where no one owns any ideas, blatantly copying entire works doesn't seem like a wrong thing to do.
Actually, being a GPL-FOSS-etc. person, I don't agree fully with you; nobody should own an idea. If only, because his/her idea is based on numerous ideas of others, who had all those ideas needed as basis. Maybe we can agree on the term owe instead? If one uses an idea, any idea, a phrase, a drawing, that isn't one's very own, one owes it to the originator, to mention one's source(s).
An oral assessment would grade presentation ability which is irrelevant to course content
I can understand your argument, if your/.-ID reflects your physical age. As someone who went through a program with almost all oral exams, in a time when a 'power point' was still a socket in the wall, delivering 220V, 50Hz, usually, I can confirm that there was nothing about 'presentation skills' in our oral exams. It would be questions by the professors from the first minute onwards, and the students' task was just to answer. And in those days, it wasn't considered politically incorrect to simply cut off excessive sentences from the students, and inject the next question instead.
I still do it at times with my own students, and find it amazingly simple, and insightful, to gauge the understanding of the student. In finals, including my own, to be frank, once too often questions appear, where the answer can be memorized; where an answer can be 'good' without containing much of knowledge and understanding. Talk to the person, and it takes less than a minute to find out if s(he) is a cheat.
No, seriously, I mean it. I don't want to be locked in, in some app-shop. Should it run Ubuntu's Maverick, I'll go for one, sure. Should it be a locked-up version, I'll skip it.
Maybe we should try a poll, so that Samsung can still liberate the hardware?
The first three posts in this discussion are - as of now - ACs. Though different from the normal 'First Piss Post'-category. They are spot on the topic. Still ACs. Why? Already fearful of being tracked? Yes, you are. Through your IP-addresses. Next year you can be tracked by having your Personalausweis in your pocket. Or in your bag. You need it, because you want to enter an official building; the Rathaus. Or doing banking business: "Guten Morgen, Frau Müller." "Uh, Sie kennen mich?" "Nein, aber Sie haben Ihren Ausweis dabei! Ich denke Sie wollen Ihren Urlaub bezahlen!?" "Woher wissen Sie das?" "Nun, als Sie hier hereinkamen, hat unsere Sicherheitssoftware gemeldet, dass Sie gerade auch im Reisebüro waren."
Because I believe in 'balance of power', and everyone to do what (s)he is doing best. We, the readers and followers, should be aware, and keep our eyes open, for stories that could be submitted. The more, the merrier. The recent case Oracle vs. Google could become a milestone in IT, (software) patent case law. Therefore, what a reader considers 'informative' or 'insightful' should be submitted. (S)he is usually not a professional, but a motivated lay person. So, well done from this side. The Slashdot staff is paid for balanced judgments of the quality, interest, integrity, relevancy of a submitted story. From on objective, professional perspective. And when I read that underlying 'story' in infoworld, it turned out to be pretty much on the side of a blabbering, non-coherent concatenation of sentences. It is the assigned task to people like timothy to separate the wheat from the chaff. He has failed.
There are ALWAYS chances of this regardless of whatever technology you use.
But of course! And of course one needs backup all the time and always. No question here. What OP (and myself) are clearly arguing about, it that the premise (promise?) was on 'as long as the hard disk does not crash, your data will be available. Reliably.' We ran all possible diagnostics up and down, on all the (different) drives on which we lost access completely; and all were 100% according to what the manufacturers' tests claimed. So, it simply is a case of 'expectations not fulfilled'. If I don't mind losing data through software (file system) errors, ext3, ntfs, even FAT, will do just as well. Snapshots are totally phantastic with ZFS, I agree. Backup is straightforward simply, volume management as well, etc. All hunky dory. But don't use it for your desktop, don't use it for your laptop, don't use it for USB drives.
USB is not only not recommended, there are warnings about it's use.
I hope someone mods you up +5, as well. As a warning.
Alas, the losses that we incurred were not based on USB drives. If there is a complete mirror of the forums, I can still retrieve the relevant messages; by the developers. 0. Never use USB drives [as you mentioned] 1. Never use a system with a single drive 2. Even a power supply that works great with Windows or Linux, can induce glitches that render a drive unreadable
This doesn't mean ZFS was 'useless'. But think of it: OpenSolaris never had reasonable support for any other file system. No, including FAT/vfat. The developers as well conceded it was kind of shoddily written, as a 'hack'; and worked most of the times. NTFS, you know, with limited success, read-only (if memory serves right; likewise ext2). So what am I supposed to do with a notebook? Where we usually have a single drive? What about a 'green' desktop with a single 'green' drive? Not encouraged. Finally, yes, on a server it makes a lot of sense. I have no qualms to support it there. Though, in the end, it's just a niche. A dangerous niche; because I rather support a filesystem that I am used to, acquainted with, instead of one here and another one there.
Yep, it is such a sad thing; I myself tried to 'sell' it as 'last filesystem mankind will need'; with atomic writes (the only reasonable solution), in-flight CRC, you name it. But it didn't deliver. That's it. Over. By now, I see the earlier phrase about filesystem and mankind from its other interpretable angle... .
Not. ZFS never made the promise of broad availability. At least not, after it was CDDL-ed. So "Oracle's inept handling...": didn't you, btw., made a typo? Wasn't it more of "SUN's inept handling..." that made ZFS not much of a good OSSitizen?
Secondly, a BSD license is such a good thing. No flamewars, but it doesn't include any non-patent-clause. And suddenly, you stand in your briefs (or even with less), if you have a company and you are sued.
Not any longer. That's what Nexenta started off with so, so very well. But the geezers of old-time SUN bosses didn't allow. At all. Maybe they were not so much of geezers after all? With the conservative clientèle of SUN, they expect no changes, 100% backward-compatibility. All those 'geeky' and 'nerdy' solutions used to be no-go-zone for those in charge. Plus: control. Think of CDDL. Stay in control. So a new package manager, it had to be. It never was possible, under SUN, to extend apt-get (I can divulge some inside info here: apt-get doesn't work with zones, e.g.. And SUN didn't want to extend apt-get for zones.)
Yes and no. Out of context, your post is almost 'Insightful'. (Or 'Common Sense'.) Within the context, though, ZFS promised no data corruption except at hard disk failure, due to atomic writes. Not wanting to delve into details, there was a devil in the details. If someone really was interested, I could do a write up; but in a nutshell, SUN (ie the developers) didn't deliver on this promise. There are a number of cases, confirmed cases, when a perfect hard disk loses data, actually all data, irrecoverably, with the hard drive being 100% okay. As I wrote, details on request (or you search the old forums of OpenSolaris on your own).
Slashdot Mirror
you can read the zero gravity toilet instructions. ... and that's exactly why I would watch 2001, and that's exactly why I would buy a DRM-ed blue ray, plus player: To read the zero gravity toilet instructions. In case I happen to sit on one tomorrow.
Of course you are right if you take my words literally, and to that extent.
However, if you care to infer the message, it is about the focus with which movies are shot. There is only that much of funding, and in these days, I'm afraid, the average investor is more concerned about resolution and brilliance (of the shots), than in the artistic quality of the undertaking. And all this 'over-technisised' appreciation of the audience will actually lead to movies being shunned because of a perceived lower technical quality, despite of potentially higher artistic quality.
I personally have overheard people who refuse to buy any non-BlueRay movie, because "Blue Ray is the future". Content seems to disappear behind technicalities, including for the consumer.
And if you please read the message of the OP, I would never want to watch 720x486 NTSC SD interlaced footage EVER again., you might understand my urge to point out what a nonsense this implies. And that one was modded +5, Informative. I was only trying to say, that my primary argument for selecting a movie is its artistic content; not its resolution.
You sound as if you knew what you are talking about, so I take it to try to answer your message.
Disparity is the most important factor in 3D perception in the human visual system
No doubt. This is why 3D 'works'.
Motion parallax however, cannot be achieved, since hidden content cannot be interpolated. It actually is unknown, eventually to both viewpoints. Even if it is known to one, depth remains unknown.
Head-mounted devices are worse, because for nobody the world doesn't turn when (s)he turns the head. You follow with the shown perspective, I guess. But from where do you get it? Think about a movie: Where do you get the information from, when the viewer turns the head?
Vergence and accommodation (focus) are secondary and always overridden by the other factors; this is a neurophysiological fact
Yes, see above. Override, though, does not mean trashed. It remains a sensory effect, that contradicts at least a distance virtually 'close' according to its disparity. -> Headaches.
How does a microlens array induce physical distance (adaptation)? You'd need a set of screens at various distances form the viewer's eyes, and using a shutter mechanism to project specific pixels from a 'credible' distance.
Physiologically, if you inhibit head/eye movement totally, the vision disappears altogether, as you probably know. So we all perform small quantities of those all the time, unconsciously. That's fine for a 2D-display (as I wrote elsewhere in this topic), because that's what we are aware of: a 2D-projection on a plane of finite, if not very limited size. Our brain 'expects' what it gets from watching a picture, or your 2D flat screen. Even a 3D-effect (compiz, e.g.) is nothing but a calculation of virtual distances and structures, projected - visibly - on a 2-dimensional screen. So our eyes get what they expect, with respect to convergence, parallax, focus, etc.
Not so, however, if you add real depth/disparity; but none of the others. Tiny, maybe subconscious, movements of eyeballs and/or your head do actually 'explore' the depth; not so in any 'disparity-is-everything'-projection system.
As long as we don't have a projection that makes appear an object in a real 3-dimensional space (what in theory a laser could do), so that the room is real, with a virtual object of real 3 dimensions projected into it, headaches will be the order of the day.
Yes, and no. Your argumentation discards a relevant fact; one that you are probably not aware of.
Black and White photos are a proper representation, or mapping, of a 3-dimensional space on a 2-dimensional plane. Adding colour adds information. The human eyes can be tricked into perceiving a rate of above 16 images per second as 'motion', and an ever higher rate as 'smooth motion'. You add colour to it, everything fine.
Over the years, this has been refined, and we can all enjoy coloured moving images without trouble.
Stereoscopy as it is being done, cannot produce a proper mapping. (I gave some initial arguments elsewhere in this thread, so I don't want to repeat myself.) This is why 3D hasn't taken off despite of very early efforts, in red/green, of some generations earlier. The problem is not one of technology, resolution, not even left/right separation. The problem is, and there is plenty of research available if you are interested, that - contrary to the mapping of 3D to 2D - two cameras - even if mounted with the proper interocular distance - cannot map the 3D-impression properly into 2 electronic channels. Therefore, it is physically/biologically impossible to regenerate the original 3D impression with lateral cameras.
1920x1080p Blu-Ray Discs are incredible. I would never want to watch 720x486 NTSC SD interlaced footage EVER again. I work in post production/special fx, so i'm a videophile.
That's one of the troubles with the world of today. Some people get their kicks just from the resolution of the image. Go to any TV-electronics parlor. People will be excited about the crisp picture, the brilliant colours. Whenever I go there, I am infinitely bored with the crappy movies. And then I go home, and watch 720x486 NTSC SD interlaced with an enormous pleasure; Bunuel, Hitchcock, Marx Brothers. Even Kubrick's 2001 is great fun, in PAL. Murnau's Nosferatu (I guess, not more than 300x200 effective resolution) sends more shivers down my spine than Kinski's remake, even if it were in 1080p.
Because it is the art; not the resolution that counts.
YMMV, though.
I'd throw all my mod points at you, if I had or could. Already the next commenter knows close to nothing about stereoscopy, but argues as if he did. You are right from a number of viewpoints (pun!, héhé):
1. At real 3D, when you move your head laterally, you can 'circle' around an object.
2. At real 3D, when you move your head laterally, objects hidden behind other objects become visible.
3. [You didn't mention this one:] Depth perception of the human eyes is done by a combination of biological effects:
- convergence of the eyeballs (like when you watch your finger and bring it closer to your own face, the eyes turn 'inward'
- adaptation of the lenses for a specific distance
The so-called '3D' that can be achieved by two cameras only fulfills one of these features: convergence of the eyeballs (by introducing a lateral offset of the two images on the projection screen).
This is why this so-called '3D' gives you some '3D-feeling', but mostly headaches; as the "3D-detection algorithm" in your brain cannot accommodate the incoming information properly; it defeats and contradicts what it has learned throughout your lifespan.
... though I am a GPL fan and against geolocation. ;)
If the wine is marked according to the grape, like Merlot or Syrah, or so, fine. If it says something that has locally specific production procedures, or stems from a specific location, I am very much for disallowing fake labels. I think the so-called 'Budweiser' has been mentioned. Except of the name, it has nothing at all to do with the original. Therefore it is misleading, over, and the consumer ought not be mislead.
Some wines have specifics, as well as cheese. Like Roquefort, which ripens under a specific situation, in a specific cave.
Jerez - actually what made it into Sherry, because the British find it difficult to pronounce Spanish names - has some specifics in its taste that cannot easily been reproduced by inputting some nondescript white wine. Gewürztraminer is another one - though I don't like white wines - in my list. Drank some in Perth, Australia once, and found it unbearable. Just plugging whatever fantasy names there are on the bottle, and off you go? No, as someone who does like red wine, it is even worse for me. A Bordeaux is distinctly different from a Beaujolais Nouveau, and when I buy one, I have a desire to enjoy a distinct taste to which I was looking forward when I bought it. And I want to have this privilege in future as well, even more so. I don't mind at all those crazy and fantastic names for wines that are sold left and right where we live, the 'Night Train', 'Red River' and likewise stuff. But when I buy a St. Emilion, I want a St. Emilion. Anything else is cheating; and I am somewhat astonished that a number of comments in here do not feel likewise. It is not much different from computers: if I buy an intel, I want neither an AMD nor a Via; and if I buy a Via, I don't want an Intel i3.
And don't come with all those silly 'double-blind tests'. Should I do likewise, and state: "If you can't distinguish the speeds of two computers, subjectively, (like with the wines), you are not entitled to obtain the type of CPU that you ordered"? When I buy an Intel Core i5 661, that's what I'm entitled to get. Irrespective of my ability to count the pins and measure the clock frequency. And when I buy a Champagne, I want explicitly a product that has been produced according to a very specific and distinct method; not some sparkling white wine with added carbon dioxide.
I suppose if they are brought up in a society where no one owns any ideas, blatantly copying entire works doesn't seem like a wrong thing to do.
Actually, being a GPL-FOSS-etc. person, I don't agree fully with you; nobody should own an idea. If only, because his/her idea is based on numerous ideas of others, who had all those ideas needed as basis.
Maybe we can agree on the term owe instead? If one uses an idea, any idea, a phrase, a drawing, that isn't one's very own, one owes it to the originator, to mention one's source(s).
An oral assessment would grade presentation ability which is irrelevant to course content
I can understand your argument, if your /.-ID reflects your physical age.
As someone who went through a program with almost all oral exams, in a time when a 'power point' was still a socket in the wall, delivering 220V, 50Hz, usually, I can confirm that there was nothing about 'presentation skills' in our oral exams. It would be questions by the professors from the first minute onwards, and the students' task was just to answer. And in those days, it wasn't considered politically incorrect to simply cut off excessive sentences from the students, and inject the next question instead.
I still do it at times with my own students, and find it amazingly simple, and insightful, to gauge the understanding of the student. In finals, including my own, to be frank, once too often questions appear, where the answer can be memorized; where an answer can be 'good' without containing much of knowledge and understanding. Talk to the person, and it takes less than a minute to find out if s(he) is a cheat.
YMMV
No, seriously, I mean it. I don't want to be locked in, in some app-shop.
Should it run Ubuntu's Maverick, I'll go for one, sure.
Should it be a locked-up version, I'll skip it.
Maybe we should try a poll, so that Samsung can still liberate the hardware?
Nevermind what the poster attempted do it.
I for one found it quite funny.
(Simply have no mod points now, so a comment should be in order.)
Thanks for turning yourself in!
Now, please, the other two!?
Where did I see Brazil?
In your own subject line. And 'high' up in the title of the post to which you replied. OMG
The first three posts in this discussion are - as of now - ACs. Though different from the normal 'First Piss Post'-category. They are spot on the topic. Still ACs. Why?
Already fearful of being tracked? Yes, you are. Through your IP-addresses.
Next year you can be tracked by having your Personalausweis in your pocket. Or in your bag. You need it, because you want to enter an official building; the Rathaus.
Or doing banking business:
"Guten Morgen, Frau Müller."
"Uh, Sie kennen mich?"
"Nein, aber Sie haben Ihren Ausweis dabei! Ich denke Sie wollen Ihren Urlaub bezahlen!?"
"Woher wissen Sie das?"
"Nun, als Sie hier hereinkamen, hat unsere Sicherheitssoftware gemeldet, dass Sie gerade auch im Reisebüro waren."
Oh, what a brave new world we weave ... .
Because I believe in 'balance of power', and everyone to do what (s)he is doing best. We, the readers and followers, should be aware, and keep our eyes open, for stories that could be submitted. The more, the merrier. The recent case Oracle vs. Google could become a milestone in IT, (software) patent case law. Therefore, what a reader considers 'informative' or 'insightful' should be submitted. (S)he is usually not a professional, but a motivated lay person. So, well done from this side.
The Slashdot staff is paid for balanced judgments of the quality, interest, integrity, relevancy of a submitted story. From on objective, professional perspective. And when I read that underlying 'story' in infoworld, it turned out to be pretty much on the side of a blabbering, non-coherent concatenation of sentences. It is the assigned task to people like timothy to separate the wheat from the chaff. He has failed.
Only since you ask.
Please, mod up the submitter. Submitting is his good right, and we should reward his efforts.
Please, mod timothy down for accepting a boring, not-even-a-story.
Please, mod the original author 'overrated', since his story should never have made it into infoworld in the first place.
Must Not Happen
There are ALWAYS chances of this regardless of whatever technology you use.
But of course! And of course one needs backup all the time and always. No question here.
What OP (and myself) are clearly arguing about, it that the premise (promise?) was on 'as long as the hard disk does not crash, your data will be available. Reliably.'
We ran all possible diagnostics up and down, on all the (different) drives on which we lost access completely; and all were 100% according to what the manufacturers' tests claimed. So, it simply is a case of 'expectations not fulfilled'. If I don't mind losing data through software (file system) errors, ext3, ntfs, even FAT, will do just as well.
Snapshots are totally phantastic with ZFS, I agree. Backup is straightforward simply, volume management as well, etc. All hunky dory.
But don't use it for your desktop, don't use it for your laptop, don't use it for USB drives.
USB is not only not recommended, there are warnings about it's use.
I hope someone mods you up +5, as well. As a warning.
Alas, the losses that we incurred were not based on USB drives. If there is a complete mirror of the forums, I can still retrieve the relevant messages; by the developers.
0. Never use USB drives [as you mentioned]
1. Never use a system with a single drive
2. Even a power supply that works great with Windows or Linux, can induce glitches that render a drive unreadable
This doesn't mean ZFS was 'useless'. But think of it: OpenSolaris never had reasonable support for any other file system. No, including FAT/vfat. The developers as well conceded it was kind of shoddily written, as a 'hack'; and worked most of the times. NTFS, you know, with limited success, read-only (if memory serves right; likewise ext2).
So what am I supposed to do with a notebook? Where we usually have a single drive? What about a 'green' desktop with a single 'green' drive? Not encouraged.
Finally, yes, on a server it makes a lot of sense. I have no qualms to support it there. Though, in the end, it's just a niche. A dangerous niche; because I rather support a filesystem that I am used to, acquainted with, instead of one here and another one there.
Yep, it is such a sad thing; I myself tried to 'sell' it as 'last filesystem mankind will need'; with atomic writes (the only reasonable solution), in-flight CRC, you name it. But it didn't deliver. That's it. Over. By now, I see the earlier phrase about filesystem and mankind from its other interpretable angle ... .
Not only Free and Open Source Software, you also beat a lefthanded iPhone hands down with TWENTY minutes of call time!
Chris?
Not. ZFS never made the promise of broad availability. At least not, after it was CDDL-ed. So "Oracle's inept handling ...": didn't you, btw., made a typo? Wasn't it more of "SUN's inept handling ..." that made ZFS not much of a good OSSitizen?
Secondly, a BSD license is such a good thing. No flamewars, but it doesn't include any non-patent-clause. And suddenly, you stand in your briefs (or even with less), if you have a company and you are sued.
But that is easy enough to fix.
Not any longer. That's what Nexenta started off with so, so very well. But the geezers of old-time SUN bosses didn't allow. At all.
Maybe they were not so much of geezers after all? With the conservative clientèle of SUN, they expect no changes, 100% backward-compatibility. All those 'geeky' and 'nerdy' solutions used to be no-go-zone for those in charge. Plus: control. Think of CDDL. Stay in control. So a new package manager, it had to be.
It never was possible, under SUN, to extend apt-get (I can divulge some inside info here: apt-get doesn't work with zones, e.g.. And SUN didn't want to extend apt-get for zones.)
If it had been GPL licensed originally then we'd have taken it and ran
or just not. Depends on the patents that the original owner holds. And on the version of GPL.
(I am a GPL-person)
Yes and no. Out of context, your post is almost 'Insightful'. (Or 'Common Sense'.)
Within the context, though, ZFS promised no data corruption except at hard disk failure, due to atomic writes. Not wanting to delve into details, there was a devil in the details. If someone really was interested, I could do a write up; but in a nutshell, SUN (ie the developers) didn't deliver on this promise. There are a number of cases, confirmed cases, when a perfect hard disk loses data, actually all data, irrecoverably, with the hard drive being 100% okay. As I wrote, details on request (or you search the old forums of OpenSolaris on your own).